skills/stanestane/game-design-skills-bundle/game-design-multiplayer-feature-audit/Gen Agent Trust Hub
game-design-multiplayer-feature-audit
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied game design content, creating an inherent surface for indirect prompt injection where malicious instructions could be embedded in the audited text.\n
- Ingestion points: User-provided descriptions of game features, live-ops, and social systems (SKILL.md).\n
- Boundary markers: The skill does not specify boundary markers or instructions to ignore embedded commands in the source text.\n
- Capability inventory: None. The skill performs no file writes, network requests, or shell command executions.\n
- Sanitization: No sanitization or input validation is prescribed for the analyzed content.
Audit Metadata